This paper introduces a novel model for image fusion that is based on a fractional-order osmosis approach. The model incorporates a definition of osmosis energy that takes into account nonlocal pixel relationships using fractional derivatives and contrast change. The proposed model was subjected to theoretical and experimental investigation. The semigroup theory was used to demonstrate the existence and uniqueness of the evolution equation solution. Additionally, the model was validated and tested using numerical experiments and compared to local image fusion methods. The findings demonstrate that the proposed model outperforms the competitive local image fusion models.

本文介绍了一种基于分数阶渗透方法的图像融合新模型。该模型结合了渗透能量的定义，该定义考虑了使用分数导数和对比度变化的非局部像素关系。所提出的模型进行了理论和实验研究。利用半群理论证明了演化方程解的存在性和唯一性。此外，该模型还使用数值实验进行了验证和测试，并与局部图像融合方法进行了比较。研究结果表明，所提出的模型优于竞争性局部图像融合模型。